October 02, 2015 3 min read

Leading Gastro Doctors give their thoughts

Publicity surrounding the FDA's approval of a stool DNA test for colorectal cancer screening made fact and fiction difficult to distinguish, according to Deborah Fisher, MD, MHS, a gastroenterologist and associate professor at Duke University. She wants to set the record straight in this guest blog.

I am a gastroenterologist and much of my research and clinical focus is colorectal cancer screening. Some of my previous comments on the new stool DNA test, Cologuard, have already been published in the New York Times as well as on MedPage Today. However, I have recently noticed a number of misleading articles in various newspapers across the country and wanted to address these, likely common, misconceptions about the new test.

First, I want to openly acknowledge the positives about Cologuard. The study in the New England Journal of Medicine examining its ability to find a colon or rectal cancer as a one-time test (compared with colonoscopy as the gold standard) was large, well-designed, and well-executed. It showed that as a one-time test Cologuard was 92% sensitive for cancer. It also showed that the false positive rate was about 13%.

The problem has arisen in how the study results are being spun. Here are a few emerging myths to debunk:

1) Cologuard is for patients who absolutely, positively want to avoid colonoscopy.


While colonoscopy is the most common test used in the U.S. to screen for colorectal cancer, it is not the only test. It is not even the "best" test. It is the most accurate test for finding cancers and polyps but that is not the same as the "best" because there are clear downsides including cost, risk of complications, access. A screening strategy of using a noninvasive test first can greatly decrease the need for colonoscopy but does not eliminate colonoscopy because positive screening tests should be followed by a colonoscopy. Therefore, a positive stool DNA test needs to be evaluated with colonoscopy or the point and potential value of screening is lost.

The fecal immunochemical test, a.k.a. FIT, is also a noninvasive colorectal cancer screening test. It has been around for decades, and in its current form is a widely available test that is included in all the U.S. colorectal cancer screening guidelines.

In fact, I will argue that it is a better test for your patient who is not completely against screening but would prefer to avoid colonoscopy if possible because 1) its false positive rate is 5% versus 13% for Cologuard and therefore it carries a lower risk of leading to an unnecessary colonoscopy, and 2) it is currently covered by all insurance carriers.

2) Cologuard will increase colorectal cancer screening rates in the U.S. population


Or perhaps a more accurate response might be, "We have no idea," since it is a new test and the only people who have used it were in a research study.

On the other hand, we have compelling data that FIT increases screening rates compared with the older guaiac-based fecal tests and among patients who were previously unscreened.

3) Cologuard is the best noninvasive test for colorectal cancer screening


And a bit of a trick question because we really do not have a Best Test at this point. Nonetheless, Exact Sciences (the test's developer) is clearly going after the screening market left over after colonoscopy takes its lion's share. In addition to my general case for no "best" test, I would argue that Cologuard has not accumulated enough data to knock FIT off its best noninvasive test perch.

Yes, Cologuard had a higher rate of diagnosing cancer in one-time testing compared with a single FIT product in a single study, but:

  • Screening is not a one-time test. FIT is meant to be repeated annually (or biennially in most screening programs outside the U.S.). We have no idea if after two or three rounds of FIT the higher sensitivity (detecting cancer among individuals with cancer) of Cologuard will remain.
  • The interval of annual FIT testing has a wealth of data support versus no data whatsoever supporting ANY testing interval for stool DNA tests (currently CMS is considering a 3-year interval).
  • In a healthy population, it is important to consider harms, and the higher rate of false positives leading to colonoscopy could translate into an increased rate of adverse events.
  • FIT is $25 versus $600 for Cologuard.